Abstract
The availability of satellite images at different spatial, temporal, radiometric and spectral resolutions is being used to monitor large areas of the various biomes in the world. The orbital image monitoring allows to detect, analyze and infer about changes in the surface of the Earth caused by natural or anthropic processes. The objective of this study was to monitor and analyze spatial and temporal evolution in land use and occupation in the Municipality of Petrolina, Pernambuco, Brazil, using surface data and Landsat-8 images for the years 2013 to 2015. Remote sensing techniques were used to estimate Normalized Difference Vegetation Index (NDVI), surface albedo (α) and Mobile Standard Deviation Index (MDSI). The technique of detection of change and the classification by decision tree based in predefined rules in the NDVI, MSDI and surface albedo were applied to infer about occupation and land use. The results showed that the vegetations indexes and degradation monitoring and change detection techniques satisfactorily identified the behavior of the surrounding vegetation in the Municipality of Petrolina, highlighting the indicators of degraded areas in the semiarid region. It was concluded that the technique of detection of change and classification by decision tree were efficient in the identification of anthropic areas during the three years studied.
Keywords:
Caatinga; albedo; NDVI; change detection; agrometeorology
